aif-plan
Pass
Audited by Gen Agent Trust Hub on Mar 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
Bashtool to execute git operations (init,branch,worktree,checkout,pull) and file system commands (mkdir,cp,cd) to manage development environments and synchronize context files. \n- [PROMPT_INJECTION]: The skill implements a hierarchical instruction system that treats repository files as authoritative overrides, creating an indirect prompt injection surface. \n - Ingestion points: Mandatory reading of
.ai-factory/skill-context/aif-plan/SKILL.md, and optional reading of.ai-factory/DESCRIPTION.md,.ai-factory/ARCHITECTURE.md, and.ai-factory/RESEARCH.md. \n - Boundary markers: No explicit delimiters or instructions to ignore embedded commands are used when interpolating file contents. \n
- Capability inventory: Includes system command execution via
Bashand task management viaTaskCreate/TaskUpdate. \n - Sanitization: Content from these files is used as-is to dictate agent behavior without validation or filtering.
Audit Metadata